Comparing and refining two novel methods for paleothermometry in speleothems - carbonate clumped isotopes vs. fluid inclusion noble gas concentrations
Final Report Abstract
Clumped isotope thermometry (Δ47) reflects the over-abundance of 13C-18O bonds in the carbonate lattice compared to that at random distribution, with low Δ47 values correlated to high temperatures. The relationship between temperature and Δ47 value was calibrated with calcite precipitated in the laboratory and is generally consistent with biogenic carbonates, whereas speleothems (cave carbonate deposits) show disequilibrium effects and deviate from the laboratory calibration. We investigated this disequilibrium using a set of modern speleothems from Bunker Cave (Germany) and adjacent caves and found deviations from the expected equilibrium Δ47 value for all samples. Δ47 offsets co-vary with δ18O offsets and yield a Δ47-δ18O relationship that is consistent with other speleothem studies. These offsets show a significant variability despite constant temperature and isotopic drip water composition and reflect variations in the kinetic isotope effect resulting from changes in super-saturation and CO2 gradient between solution and cave atmosphere. A comparison with a common test for kinetic isotope fractionation in speleothems (the so-called ‘Hendy test’) shows high Δ47 offsets to be related to a positive Hendy test (indicating kinetic isotope fractionation) and low Δ47 offsets to a negative Hendy test. Clumped isotopes appear to be more sensitive to kinetic isotope fractionation than the Hendy test. Thus, Δ47 measurements are a new and sensitive tool for investigating kinetic isotope effects in speleothems and possibly other carbonates. We applied the clumped isotopes method to calcite from Devils Hole, to check if it was precipitated at equilibrium, as expected from its very slow growth and low super saturation. Δ47 results show constant values for samples from 27-179 ka in agreement with modern instrumental water temperature (33-34 °C). Δ47 values corresponding to the inorganic laboratory calibration confirm Devils Hole calcite to reflect equilibrium precipitation. As a consequence the commonly used value for the oxygen isotope fractionation in carbonate formation has to be revised (1.5 ‰ difference between Devils Hole calcite and laboratory data). Using clumped isotopes we also found carbonates at low temperatures (~0 °C; cryogenic cave carbonates) that were precipitated slowly and don’t show offsets compared to the Δ47–temperature calibration. Together with the Devils Hole data they could be used to establish a revised oxygen isotope fractionation and its true temperature dependence. The close correlation between Δ47 and δ18O offsets can be used to correct for disequilibrium in speleothems and to quantitatively reconstruct regional paleoclimates. This application requires one independent parameter: T for water δ18Ow reconstruction and the paleo-fluid δ18Ow value for T reconstruction. We reconstructed the δ18Ow of paleo-water from Holocene and Late Pleistocene stalagmites from Bunker Cave using temperature constraints from independent studies and the clumped isotope method to correct for disequilibrium. The paleo-water δ18Ow values are more negative (~1.5 ‰) during colder climatic periods and similar to modern values during Interglacial conditions, in agreement with other archives. A temperature reconstruction, that uses Holocene stalagmites and a fluid δ18Ow composition inferred from a lake record, results in temperatures consistent with the previously determined noble gas temperatures from speleothem fluid inclusions and shows the potential of the clumped isotope approach for cross-calibration of different proxies in particular and for the paleoclimate reconstruction in general.
